OpenAI Launches First Day as AI Industry Grows in Cybersecurity

OpenAI was launched on Monday Ripea cybersecurity project that aims to help developers and security teams identify vulnerabilities, ensure remediation, and protect software quickly using artificial intelligence.

The announcement underscores the big shift the AI ​​industry is increasingly making to cybersecurity as leading brands advance in code analysis, finding software vulnerabilities, and automating technology operations.

In a post at X, OpenAI CEO Sam Altman called Daybreak “an effort to accelerate cyber security with continuous security software.”

“AI is already great and is about to make a breakthrough in cybersecurity; we want to start working with as many companies as we can now to help them continuously protect themselves,” Altman said.

According to OpenAI, Daybreak combines the company’s AI models with Codex, its security monitoring system, to help security teams review code, analyze dependencies, model threats, verify patches, and investigate unknown behavior. The company said its goal is to reduce the time between identifying a vulnerability and fixing it.

OpenAI did not immediately respond to a request for comment Decrypt.

Daybreak comes as cybersecurity researchers and industry experts warn of the threat of AI-driven cyberattacks after they launch Claude Mythos last month. Using Mythos, Firefox browser developer Mozilla said they were able to find 271 unknown problems in the browser.

“AI can now help defenders think deeply about codebases, identify subtle vulnerabilities, validate changes, analyze unknown behavior, and transition from discovery to remediation faster,” OpenAI said in a statement. “Because these same powers can be misused, the Daybreak pairs have developed defensive capabilities with confidence, assurance, security, and accountability.”

The announcement also comes as major AI companies are increasingly marketing their own cybersecurity and software engineering services. Anthropic’s OpenAI competition has also increased to be sold Claude’s examples of documents and services related to security as competition increases among AI companies seeking commercial customers.

While the experts remain division on the scale of the threat of AI, researchers and government agencies have warned that advanced forms of AI may accelerate. cyberattacks by helping hackers to analyze vulnerabilities, create malware, and destroy the environment. At the same time, Google researchers recently said that major languages ​​are developing well awareness and exploit software vulnerabilities that security scanners often miss.

OpenAI says it plans to work with governments and industry partners before deploying more cyber-capable AI models, as regulators and national security officials test. analyze before advanced forms of AI are introduced to humans.

“Dawn is the first ray of sunlight in the morning,” OpenAI wrote. “In cyber security, it means seeing a threat early, taking action quickly, and helping to make the program resilient in the process.”